The mediating role of international student satisfaction in the influence of higher education service quality on institutional reputation in Taiwan

Abstract

The number of international students applying to universities in Taiwan is increasing annually. Moreover, universities’ choice for international students is also increasing with support from the Ministry of Education of Taiwan. Therefore, it is crucial to understand the factors that influence student choice. This study investigates the effects of Taiwan’s higher education service quality on international student satisfaction and institutional reputation. We collected data from 197 international students studying at 33 national and private universities in Taiwan to test the study’s proposed model. This study uses a quantitative approach to examine the proposed hypotheses, and causal research design is used to elucidate the cause-and-effect relationship between constructs. The non-academic aspect of service quality is the most influential variable in student satisfaction, and student satisfaction strongly affects institutional reputation. Student satisfaction mediates the relationship between the academic and non-academic aspects of service quality and institutional reputation. This study’s findings could help higher education institutions in Taiwan assess and improve their service quality, which would help them attract international students.

Keywords

Service quality student satisfaction institutional reputation higher education Taiwan

Introduction

In this era of globalization, nations worldwide need to enhance their citizens’ skills, knowledge, and capability to compete globally. Higher education institutions (HEIs) are becoming more aware of the effects of globalization. In reality, the need for international work experience and partially or fully acquired education abroad is increasing.¹

The educational system and cultural values of Taiwan are similar to those of other East Asian countries. The internationalization of Taiwan’s educational system has transformed the educational system of Taiwan significantly over the last two decades.² The number of international students choosing Taiwan for their higher studies has continuously risen. The number of international students in Taiwan has continuously increased from 111,000 in 2015 to 132,000 in 2019.³

International students often struggle with choosing a suitable university, especially during the application process’s early stages. The process of selecting a university that matches the desired major is time-consuming and challenging. However, this critical decision affects the next few years of their life. Why do international students invest so much time in university selection? What influences their choice? These concerns are closely related to the students, and they affect university policymaking and management, which are ultimately related to economic benefits. Therefore, colleges and universities attracting international students is a vital issue for local governments and university administrators. Among other concerns, students focus on world-ranking, recognition, and employment opportunities after graduation. Other factors related to study and campus life are decisive factors for students in their choice of universities. For example, factors such as the school’s geographical location, environment, tuition fees, and living expenses are non-academic factors.

The number of international students in Taiwanese HEIs is increasing.⁴ The increasing mobility of international students toward Taiwan’s higher education provides opportunities for both students and universities. Students have a pervasive range of choices, and institutions face increased competition.⁵ Competition among universities to recruit international students leads to a focus on student satisfaction.⁶ Since universities compete to attract the best and brightest students, they should assess service quality and student satisfaction.⁷

Taiwanese educational institutions are facing student shortage due to low birthrate.⁸ One in three Taiwanese colleges is expected to cease operations by 2021 because of the shortage of local students,⁹ leading to intense competition to attract international students. However, Taiwan has recently started being favored by international students due to its cost-effective price, the popularity of doctoral teaching degrees in universities, and the relatively international education mode. Therefore, many universities focus on enhancing student satisfaction and reputation to attract the best international students.⁶ The number of international students registering at Taiwanese universities or colleges has increased annually, and it accounted for 10% of the total number of university and college students in 2018.¹⁰ This study endeavors to build a framework empirically to investigate the effect of Taiwan’s higher education service quality on student satisfaction and institutional reputation.

The study aims to explore the extent to which service quality dimensions (academic and non-academic) influence student satisfaction and general reputation, examine whether student satisfaction mediates the relationship between service quality (academic and non-academic aspects) and reputation, and analyze whether student satisfaction influences the reputation of an institution or university. By studying the widespread impact of service quality on Taiwanese universities, empirical data will be provided for practitioners to plan for enrollment in universities better and improve the competitiveness of institutions and the domestic higher education system.

Literature review

Service quality, student satisfaction, and reputation

In the past two decades, service quality has been extensively researched. Willms et al. proposed that the nature of service quality refers to performance and opportunity quality, which is mainly controlled by the provider.¹¹ Quality can also refer to and is measured as belief statements or attribute performance.¹² Parasuraman et al. noted the different dimensions of service quality and proposed the SERVQUAL framework.¹³ SERVQUAL is a service quality gap model, which has been used to assess service quality in different domains such as internet marketing,¹⁴ hospitality,¹⁵ restaurants,¹⁶ banking,^17,18 commercial hospitals,¹⁹ insurance,²⁰ and higher education.^21,22

Some studies have investigated the relationship between service quality and customer satisfaction. They indicated that the multidimensional service quality scale is positively and significantly associated with customer satisfaction in banking,²³ telecommunication,²⁴ the auto repair service industry,²⁵ etc. Although the industries’ nature varied, the conclusions corroborated that service quality and customer satisfaction are crucial components of the service industry. HEIs have been emphasizing service quality to gain a competitive advantage.²⁶ Higher education focuses on what society values in the graduates’ skills and competencies, as well as graduates’ educational experience.²⁷ Moreover, since higher education lies in the service industry field, universities and other related HEIs need to do their best to meet students’ service quality standards.²⁷ Govender et al. stated that service quality in higher education measures how well HEIs conform to their customers’ needs and expectations.²⁸

Maintaining satisfied customers is the key to developing competitive advantages in the current market. A deeper understanding of how to attain a certain level of customer satisfaction and awareness of the vital roles that measurement issues play is essential for a competitive advantage.²⁹ Some marketing researchers propose that satisfaction results from the high-quality performance.^30
–32 Spreng and Chiou indicate that the interdependent relationship between service quality and customer loyalty, which is essential for any enterprise, is one of the most critical and frequently-tested ideas in terms of the concept and practice of marketing.³³ The perception of service quality is an important variable affecting consumer evaluation on satisfaction.^34
–36 For example, in the tourism industry, causality can be found between suppliers’ performance, customer satisfaction, and one specific institution or enterprise’s success. In the early years, Oliver defined satisfaction as “a function of an initial standard and some perceived discrepancy from the initial reference point.”³⁷ He also quoted, “customer’s purchasing intention depends on his/her level of satisfaction.” Oliver further identified satisfaction, “the consumer’s fulfillment response, the degree to which the level of fulfillment is pleasant or unpleasant” (p.28) as an “advanced conceptualization.”³⁷ Satisfaction is an emotional state of mind after an immediate connection. Satisfaction may relate to customers’ socio-psychological status, such as tourists’ emotional attachment toward a destination. It could also relate to environmental impacts (for example, climate change, mobility of social network) beyond the host’s constraints, and all the providers and suppliers involved. Moreover, customer satisfaction has been recognized as a key element of success and survival for an organization by Athanassopoulos et al.³⁸ However, few studies discuss the relationship between service quality and customer satisfaction in the context of HEIs, where students play the role of customers.

Abdullah presented Higher Education PERFormance (HEdPERF), a performance-based measuring scale that attempts to determine service quality factors in the higher education sector.³⁹ HEdPERF uses a five-dimensional scale to measure service quality, including academic aspects (AA), non-academic aspects (NA), program issues, access, and reputation. Ali et al. stated that HEdPERF is a broad scale in which a broad range of service attributes is covered in the context of universities, and that academic and non-academic aspects are the essential dimensions that influence international student satisfaction.⁴⁰

Academic aspects (AA) are described as academic responsibilities, which include teaching, conducting examinations, and assessing.⁴¹ Academic staff is evaluated for professionalism, competence, performance, and willingness to assist students. Greimel and Geyer observed that students need teachers to clearly and comprehensibly explain new learning content and provide answers.⁴²

Non-academic aspects (NA) include social life quality (i.e., student and psychological health counseling), admissions, financial services, cafeterias, library services, and other services not directly related to teaching and learning activities.⁴¹ Non-academic staff provides these services. The capability and willingness of the administrative staff and other service providers regarding respect for the other party (customer/student), equal treatment, and protection of the confidentiality of information are emphasized.³⁹

Student satisfaction (SS) is a cognitive or affective reaction to a single or prolonged set of services that students encounter.⁴⁰ Service quality also produces customer satisfaction.⁴³ According to Helgesen and Nesset, student satisfaction is an attitude resulting from assessing student experience regarding service quality.⁴⁴ Students who feel satisfied tend to commit to a brand and repurchase products or services. Accordingly, customer (student) loyalty has rapidly become the marketplace currency of the 21st century and, consequently, a marker of an educational institution’s value.⁵

Reputation (RP) refers to the positivity or negativity of an institution’s image, which comes to the consumer’s mind when viewing a brand.⁴⁵ Verčič et al. proposed that initial image-formation is a continuous, gradual process, in which the image becomes meaningful to the client based on their thoughts, emotions, and previous experiences regarding an object or institution.⁴⁶ Organizational reputation is a type of belief, namely, the public’s emotional evaluation of an organization.⁴⁷ Generally speaking, reputation is an overall expression of general evaluations made over time, to the extent that stakeholders perceive trustworthiness, value, and respect when mentioning an organization.⁴⁸ This kind of shared perception of feelings by the public toward a particular organization may bring practical components, including high/favorable and low/unfavorable ones. Researchers state that social complexity and causal ambiguity are involved in reputation formation and provide the appearance of value and imitability, which are ultimately considered intangible assets.^49,50 Accordingly, a high reputation can offer certain advantages to an organization, such as obtaining better resources and hiring more competent staff, ultimately resulting in greater profitability opportunities.^51
–53 Favorable reputations motivate organizational achievements,^53,54 and various factors affect an organization’s reputation, including implementation, competition, affiliation, and industry.^53,55 However, some scholars argue that a high reputation could have adverse effects, which may be a burden under specific circumstances, including severe automobile recalls⁵⁶ and low financial performance.⁵⁷ The commonly discussed reputation is the corporate one. A corporate reputation is primarily associated with identity and image.⁵⁸ According to Pruzan, it represents a comprehensive view of a company’s different stakeholders.⁵⁹ In this study, the focus is on institutional reputation (generally referring to HEIs). Since students choose colleges after assessing them, they must choose based on expectations, which are affected by institutional reputation.

Furthermore, a university’s image is a perception of services, both communicative and cognitive.⁴⁰ It is considered a variable that has a strong relationship with satisfaction.⁵⁸

Interrelationship between service quality (SQ) and student satisfaction (SS)

Lai⁶⁰ and Subrahmanyam⁶¹ found a strong relationship between service quality and customer (student) satisfaction. Ali et al. stated that student satisfaction is significantly affected by effective communication between students and faculty.⁴⁰ Additionally, the academic staff’s quality of service and their interaction with students are related to retention rates and student satisfaction.⁶² Fernandes et al. demonstrated that students who are satisfied with university facilities and services are more likely to be loyal to the university.⁶³

Interrelationship between service quality and reputation

Research has proven a direct correlation between service quality and reputation.⁶⁴ Lestari et al. showed that students who receive better service quality from an institution tend to build positive images and perceptions of that institution.⁶⁵

Interrelationship between student satisfaction (SS) and reputation (RP)

Brand image is essential in the educational sector as it is in other categories of service organizations.²⁶ Kunanusorn and Puttawong proposed that institutional image might be derived from consumer satisfaction.⁶⁶ Hence, a favorable institutional image may lead to substantial student satisfaction.⁴⁰

Mediating effects between student satisfaction (SS), service quality (SQ), and reputation (RP)

Better service quality yields higher customer satisfaction, which leads to the formation of a good reputation.⁶⁰ Ali et al. revealed that student satisfaction is influenced by many aspects, including academic and non-academic aspects and effective communication between students and faculty.⁴⁰ Luo et al. stressed that student satisfaction is essential because it is the only performance indicator of service quality for service providers in higher education.⁶⁷ In other words, the service quality of HEIs in terms of faculty performance, advisory staff performance, learning activities, and library services determine student satisfaction. Similarly, some researchers have shown that service quality also affects a company’s image or reputation.^65,66 In this vein, Ali et al. found that student satisfaction affects university image significantly, and that satisfied students perceive the institutional image positively.⁴⁰

Methodology

This study uses a quantitative method to examine the hypotheses because the data are observable and measurable. We distributed questionnaires (in English) to collect data from 197 international students at 33 national and private universities in Taiwan to measure and explain the relationship between constructs (independent, mediating, and dependent variables). To choose the international students, we first contacted the International Affairs Office or the primary office to recruit international students by email or phone. Each Taiwanese university has an international student president or chief officer, the central point of communication for events and activities. We distributed the questionnaires through the chief officers of each university.

Figure 1 shows the research framework for this study. It is one dependent variable (Reputation—RP), two independent variables (Academic Aspect service—AA and Non-academic Aspect service—NA), and one mediating variable (Student Satisfaction—SS). Based on the discussion presented in the literature, five main hypotheses and two mediating hypotheses were developed for testing (Figure 1):

H1: Academic Aspects (AA) of Service Quality (SQ) influence Reputation (RP) significantly.

H2: Academic Aspects (AA) of Service Quality (SQ) influence Student Satisfaction (SS) significantly.

H3: Non-Academic Aspects (NA) of Service Quality (SQ) influence Student Satisfaction (SS) significantly.

H4: Non-Academic Aspects (NA) of Service Quality (SQ) influence Reputation (RP) significantly.

H5: Student Satisfaction (SS) influences Reputation (RP) significantly.

H6a: Student Satisfaction (SS) mediates the relationship between Academic Aspects (AA) of Service Quality (SQ) and Reputation (RP) significantly.

H6b: Student Satisfaction (SS) mediates the relationship between Non-Academic Aspects (NA) of Service Quality (SQ) and Reputation (RP) significantly.

Figure 1.

Research framework.

The interrelationship between service quality, customer satisfaction, and corporate image has been investigated by several scholars mentioned in the previous section. We chose Ali et al.⁴⁰ Furthermore, Nguyen and LeBlanc⁶⁸ to model our framework and construct our research model. The questionnaire consisted of two sections. The first described demographics and the general information of respondents. The second included the instrument of the constructs adopted by Ali et al.⁴⁰ and Nguyen and LeBlanc.⁶⁸ Four items measured the academic aspects, and four items measured the non-academic aspects. Four items measured reputation, and eight items measured student satisfaction. The satisfaction measure corroborates the works of Teas³⁵ and Hunt,³⁶ which are related to customer satisfaction and service quality. We used a 5-point Likert scale ranging from strongly disagree (1) to strongly agree (5). For the questionnaire, see the Appendix 1. The study population is comprised of international students studying in Taiwan.

Research findings

The sample consisted of 197 international students, of which 51.8% were female, and 48.2% were male. They were from 33 universities in Taiwan. As shown in Table 1, almost all were single (89.8%). 83.3% of the sample was aged between 20 and 30. 58.4% were pursuing a master’s degree, 23.4% were studying for a bachelor’s degree, and 18.2% worked on their Ph.D. The students were funded by universities and organizations (53.3%), parents (30.5%) or their government (16.2%). The majority of students were enrolled in the first, second, or third semester (58.9%), and the respondents were from 23 different countries.

Table 1.

Demographic of respondents.

Profile	Categories	Frequency	Percentage (%)
Gender	Female	102	51.8
	Male	95	48.2
	Total	197	100
Age	Less than 20	10	5.1
	20–24	70	42.1
	25–30	81	41.1
	Above 30	43	11.7
	Total	197	100
Marital Status	Married	20	10.2
	Single	177	89.8
	Total	197	100
Program Level	Bachelor	46	23.4
	Master	115	58.4
	PhD	36	18.2
	Total	197	100
Source of Funding	Government	32	16.2
	Parents/Self	60	30.5
	University/Organization	105	53.3
	Total	197	100
Semester	1–3	116	58.9
	4–6	62	31.5
	7–9	13	6.6
	10–14	6	3.0
	Total	197	100

Exploratory Factor Analysis (EFA) was used to reduce unnecessary items and increase the data’s validity. Satisfactory Kaiser-Meyer-Olkin (KMO) and Bartlett’s Test of sphericity scores were required before proceeding to EFA (Table 2). The factor loading should be 0.50 or greater.⁶⁹ SS2 and SS6 for student satisfaction and RP5 for reputation had low factor loading values (less than 0.50). With these three variables deleted, the other factor loadings were higher than 0.50, indicating the reliability of the study’s instrumentation (Table 2). The KMO score was good (0.935). The p-value was 0.000, and the total variance extracted was 73.07%. Table 3 shows the EFA results (component, mean, SD, variance extracted explained, and KMO) for each item, as well as the total KMO and variance extracted (Table 3).

Table 2.

Deleted questionnaire items and reason.

Variables	Items	Deleted Questionnaire Item	Reason
SS	SS2	The university offers programs in a flexible manner	Bad factor loading with other items
	SS6	I will continue at the same university	Bad factor loading with other items
RP	RP5	The image projected by my university influences the value of my degree	Bad factor loading with other items

Table 3.

Exploratory factor analysis.

Variables	Component				Mean	Std.	Variance Extracted	KMO
	1	2	3	4		Deviation	Explained
AA1				0.696	4.18	0.781	62.49%	0.777
AA2				0.795	4.08	0.735
AA3				0.706	4.08	0.841
AA4				0.647	4.26	0.770
NA1		0.730			3.98	0.878	75.69%	0.841
NA2		0.800			4.08	0.897
NA3		0.789			3.93	0.921
NA4		0.781			3.92	1.037
SS1	0.534				3.92	0.939	75.96%	0.917
SS3	0.829				3.94	0.983
SS4	0.771				3.85	1.019
SS5	0.838				3.87	1.059
SS7	0.721				3.77	1.086
SS8	0.619				4.10	1.035
RP1			0.812		4.07	0.895	71.75%	0.809
RP2			0.568		3.73	0.940
RP3			0.783		4.23	0.804
RP4			0.760		4.04	0.900
Total							73.07%	0.935

Extraction Method: Principal Component Analysis

Rotation Method: Varimax with Kaiser Normalization

The reliability test was used to ensure the consistency of the measurement items over time. Cronbach’s alpha coefficient was used to measure the study’s reliability. If Cronbach’s alpha is between 0.60 and 0.80, the instrument is reliable; if it is higher than 0.80, the instrument is highly reliable.⁷⁰ For this study, since Cronbach’s alpha was 0.947 for all questionnaire items (18 items), the instrument was highly reliable (Table 4). The reliability test was also conducted separately for each variable. Cronbach’s alpha for each variable ranged from 0.798 to 0.935, which indicates that each variable was highly reliable and acceptable (Table 4).

Table 4.

Reliability of each variable (N = 197).

No	Variables	Cronbach’s Alpha (α)	Number of Items
1	AA	0.798	4
2	NA	0.890	4
3	SS	0.935	6
4	RP	0.864	4

Confirmatory factor analysis (CFA) is used to study the relationship between observed variables and a set of continuous latent variables when the observed variables are categorical.⁷¹ The factor structure was explored to improve the overall fit of the model. This procedure involves ensuring unnecessary items are rejected due to their lack of correlation with other similar items. In this study, CFA was performed to examine the relationship between the items and their respective latent variables using the Analysis of Moment Structures 22 software (Figure 2).

Figure 2.

Confirmatory factor analysis (CFA) framework.

Table 5 presents the CFA results. All the values met the required criteria, indicating that the model has a good level of fit: X²/df = 1.91, Goodness of Fit Index (GFI) = 0.884, Comparative Fit Index (CFI) = 0.955, Tucker-Lewis Index (TLI) = 0.95, and Root Mean Square Error of Approximation (RMSEA) = 0.06.

Table 5.

CFA goodness of fit test (N = 197).

GOF Index	Acceptable value	CFA Model	References
x² (chi-square)		247.37
df (Degree of freedom)		129
x²/df	<3	1.91	Hair et al.⁷²
GFI	>0.8	0.88	Baumgartner and Homburg⁷³
CFI	>0.9	0.96	Kline⁷⁴
TLI	>0.9	0.94	Brown⁷⁵
RMSEA	≤0.08	0.06	MacCallum et al.⁷⁶

Calculating CFA as part of the Structural Equation Model (SEM) requires establishing convergent and discriminant validity and the instrument’s reliability. Anderson and Gerbing suggest using three standard indices to deal with convergent validity, including individual item reliability, composite reliability (CR), and the average variance extracted (AVE).⁷⁷ AVE is a measure of the error-free variance of a set of items; CR is a measure of reliability and internal consistency based on the square of the total of factor loadings for a construct; other measures used to establish validity and reliability include maximum shared variance (MSV) and maximum inter-construct correlation (Max r).⁶⁹

In this study, all the AVE values were ≤ 0.50, and the MSV value was more significant than the AVE, which meant that the four variables indicated strong discriminant validity (Table 6). Discriminant validity is tested to identify the extent to which the construct does not correlate with other measures. These are objective tests and are based on numerical scores indicating how well the construct conforms to theoretical expectations. The CR values of each latent variable were between 0.800 and 0.936, indicating strong composite reliability (Table 6).⁷⁸ Hence, the variables AA, NA, SS, and RP have strong reliability and convergent validity.

Table 6.

Test of discriminant validity, convergent validity, and composite reliability.

	CR	AVE	MSV	Max r	RP	AA	NA	SS
RP	0.873	0.634	0.604	0.777	0.796
AA	0.800	0.500	0.482	0.694	0.627	0.707
NA	0.893	0.677	0.491	0.701	0.643	0.683	0.823
SS	0.938	0.721	0.604	0.777	0.777	0.694	0.701	0.849

Note: CR > 0.7; AVE > 0.5; MSV < AVE; √AVE > Max r, is boldface diagonal

Table 7 presents a comprehensive summary of all the reliability and validity results from EFA loading, CFA loading, SEM loading, Mean, SD, Cronbach’s alpha, KMO, CR, and AVE. All the measures used in this study indicated an excellent and high variance, which means the instruments used in this study were valid and reliable.

Table 7.

Summary of reliability and validity.

Variables	Items	EFA	CFA	SEM	Mean	SD	Cronbach’s Alpha (α)	KMO
Variables	Items	loading	loading	loading	Mean	SD	Cronbach’s Alpha (α)	KMO
Minimum
Acceptable		>0.5	>0.5	>0.5			>0.6	>0.5
Value
AA	AA1	0.696	0.68	0.68	4.18	0.781	0.95	0.777
	AA2	0.795	0.71	0.71	4.08	0.735
	AA3	0.706	0.72	0.72	4.08	0.841
	AA4	0.647	0.71	0.71	4.26	0.770
NA	NA1	0.73	0.83	0.83	3.98	0.878	0.798	0.841
	NA2	0.8	0.83	0.83	4.08	0.897
	NA3	0.789	0.85	0.85	3.93	0.921
	NA4	0.781	0.77	0.77	3.92	1.037
SS	SS1	0.534	0.73	0.73	3.92	0.939	0.921	0.917
	SS3	0.829	0.89	0.89	3.94	0.983
	SS4	0.771	0.95	0.95	3.85	1.019
	SS5	0.838	0.94	0.94	3.87	1.059
	SS7	0.721	0.87	0.87	3.77	1.086
	SS8	0.619	0.68	0.68	4.1	1.035
RP	RP1	0.812	0.83	0.83	4.07	0.895	0.864	0.809
	RP2	0.568	0.66	0.66	3.73	0.940
	RP3	0.783	0.82	0.82	4.23	0.804
	RP4	0.76	0.86	0.86	4.04	0.900
Total							0.954	0.935

* Items details for each construct:

AA1. Professors have the knowledge to answer my questions relating to the course content

AA2. Professors deal in a courteous manner

AA3. Professors show a sincere interest to solve my problem

AA4. Professors show positive attitude toward students

NA1. Academic staff show sincere positive work attitude toward students

NA2. Academic staff is willing to respond to my request for assistance

NA3. Academic staff is knowledgeable to respond to my request

NA4. Students are treated equally by academic staff

SS1. The university offers quality programs

SS3. I am satisfied with my decision to attend this university

SS4. My choice to enrol in this university is a wise one

SS5. I am happy with my decision to enrol in this university

SS7. I will recommend this university to my friends and family

SS8. I will keep in touch with my advisor or university after I graduate

RP1. This university has a prestigious image

RP2. This university’s graduates are easily employable

RP3. This university has a good image

RP4. I have heard positive things about this university

According to Guh et al., SEM is used to analyze the measurement model, estimate the structural model, and test the proposed research hypotheses.⁷⁹ Figure 3 shows the SEM testing result.

Figure 3.

The result of structural equation model (SEM) testing.

Table 8 presents the results of the fit indices of the proposed model. X²/df = 1.92, and as it is a value between 2 and 1, it is an acceptable fit with p-value = 0.00. Goodness of Fit Index (GFI) = 0.884, Comparative Fit Index (CFI) = 0.955, Adjusted Goodness of Fit Index (AGFI) = 0.846, and Root Mean Square Error of Approximation (RMSEA) = 0.06. Table 8 shows that each point met the minimum acceptable value requirement.

Table 8.

SEM goodness-of-fit statistic.

GOF Index	Acceptable Value	SEM Model
x²/df	<3	1.92
Probability	P < 0.05	.000
GFI	>0.8	.884
CFI	>0.9	.955
AGFI	>0.8	.846
RMSEA	≤0.08	.068

Table 9 and Figure 4 present the result of the proposed hypotheses. They show that H1 and H3 were rejected. H1 was rejected because the path from AA to RP was not significant (p-value = 0.253) with β = 0.11. Nevertheless, the path from AA to SS showed strong influence with β = 0.40***, and NA positively influenced SS (β = 0.43***). H4 was rejected because β = 0.15 with p-value = 0.104. Finally, the path from SS to RP was supported because β = 0.59*** Therefore, H5 was supported.

Table 9.

Result of the hypotheses based on standardized estimates.

Hypotheses	Path	β	S.E	P-value	Result
H1	AA ⇒ RP	0.11	0.137	0.253	Rejected
H2	AA ⇒ SS	0.40***	0.132	0.000	Supported
H3	NA ⇒ SS	0.43***	0.083	0.000	Supported
H4	NA ⇒ RP	0.15	0.085	0.104	Rejected
H5	SS ⇒ RP	0.59***	0.108	0.000	Supported

* P < 0.05, **P < 0.01, ***P < 0.00

Figure 4.

Standardized results (β) of structural equation model (SEM).

To address this paper’s research questions, we examined the level of the mediating effect of each proposed mediator and checked whether it was a partial or complete mediator. Bootstrapping was conducted using Amos 22. When bootstrapping, the direct, indirect, and total effects were tested. This study’s bootstrap samples were set to 5000, and the bias-corrected bootstrap was set to 95% confidence interval (CI). Figure 5 and Table 10 present the result of the mediation analysis.

Figure 5.

SEM with mediation and without mediation.

Table 10.

Summary of mediation for the model using bootstrapping.

		Indirect (ab)	Direct (c’)	Total (c)	Mediation
H6a	AA → SS → RP	0.239**	0.113 (NS)	0.352**	Supported
H6b	NA →SS → RP	0.253*	0.150	0.403*	Supported

*P < 0.05, **P < 0.01, ***P < 0.00, NS = NOT Significant.

Table 11 shows that the first three steps are supported. The results shown in this table indicate a difference between total effect (β = 0.63***) and direct effect (β = 0.17NS).

Table 11.

Testing the mediating effect of SS on predictor AA and outcome RP.

Testing step	Predictor	Mediator	Outcome	Std. β	Mediation of SS

Step 1	AA	No	RP	0.63***	Possible

Step 2	AA	No	SS	0.69***	Possible
Step 3	SS	No	RP	0.78***	Possible
Step 4	AA	SS	RP	0.69***	Accepted
				0.66***
				0.17

* P < 0.05; **P < 0.01; ***P < .001; Std β = standardized beta coefficient

Figure 6 shows the graphic representation of mediation. Table 12 shows in detail the effect size of SS on AA and RP. The indirect mediation effect is significant (β = 45***). This table also indicates that the indirect to total effect ratio is β = 0.73, and the ratio of R² _indirect/R² _total = 0.925. This indicates that 92.5% of the explained variance in the relationship between AA and RP (with SS as the mediator) is due to SS’s mediating effect and is thus statistically significant. Moreover, R² _indirect/R² _total = 0.925 indicates that SS is a mediator between AA and RP. Additionally, it indicates complete mediation.

Figure 6.

Mediation analysis for AA, SS, and RP.

Table 12.

The effect size measure of SS on AA and RP.

Type of mediation	β	R²
Total	0.63***	0.40
Direct	0.17 NS	0.01
Indirect	0.46***	0.39
Indirect to Total Ratio	0.71	0.925

* P < 0.05, **P < 0.01, ***P < 0.001; NS = Not Significant

Table 13 shows that the first three steps are supported. The results from Table 13 indicate a difference between total effect (β = 0.644***) and direct effect. (β = 0.195*).

Table 13.

Testing the mediating effect of SS on predictor NA and outcome RP.

Testing step	Predictor	Mediator	Outcome	Std. β	Mediation of SS

Step 1	NA	No	RP	0.64***	Possible
Step 2	NA	No	SS	0.70***	Possible
Step 3	SS	No	RP	0.78***	Possible
Step 4	NA	SS	RP	0.70***	Accepted

* P < 0.05, **P < 0.01, ***P < .001; Stdβ = standardized beta coefficient

Figure 7 shows the graphic representation of mediation. Table 14 shows in detail the effect size of SS on NA and RP. The result indicates that the indirect effect is β = 0.45*** and that the ratio of indirect to total effect is β = 0.83. Moreover, the ratio of R² _indirect/R² _total = 0.90, indicating that 90% of the explained variance in the relationship between NA and RP (with SS as mediator) is due to SS’s mediating effect and is thus statistically significant. Further, R² _indirect/R² _total = 0.90 indicates that SS is a mediator between NA and RP. Therefore, there is partial mediation.

Figure 7.

Mediation analysis for NA, SS, and RP.

Table 14.

The effect size measure of the whole model.

Type of mediation	β	R²
Total	0.64***	0.41
Direct	0.19*	0.04
Indirect	0.45***	0.37
Indirect to Total Ratio	0.70	0.90

* P < 0.05, **P < 0.01, ***P < 0.001

The mediating test of SS indicates complete and partial mediation. Table 15 shows that H6a is supported. The mediation is complete because AA is no longer significant when SS is controlled. The total effect between AA and RP is 0.63***, and after mediation, the direct effect becomes 0.17. Meanwhile, the total effect between NA and RP is 0.64***. After mediation, the direct effect becomes 0.195*. Hence, H6b is supported and indicated as partial mediation because NA, albeit weak, is still significant.

Table 15.

Summary results of post hoc mediation analysis.

		Indirect (ab)	Direct (c’)	Total (c)	Mediation
H6a	AA → SS → RP	.46***	0.17 (NS)	0.63***	Full Mediation
H6b	NA → SS → RP	0.45***	.19*	0.64***	Partial Mediation

*P < 0.05, **P < 0.01, ***P < 0.00, NS = NOT Significant.

To sum up, the following are the summaries of all the results of the hypotheses. They are shown in Table 16.

Table 16.

Summary table of all hypotheses result.

Items	Hypotheses	Result
H1	AA significantly influences RP	Rejected
H2	AA significantly influences SS	Supported
H3	NA significantly influences SS	Supported
H4	NA significantly influences RP	Rejected
H5	SS significantly influences RP	Supported
H6a	SS mediates the relationship between AA and RP	Supported
H6b	SS mediates the relationship between NA and RP	Supported

H1 is rejected

Academic Aspects positively and insignificantly affect Reputation (β = 0.11, p < 0.253).

H2 is supported

Academic Aspects positively and significantly affect Student Satisfaction (β = 0.40, p < 0.000).

H3 is supported

Non-Academic Aspects positively and significantly affect Student Satisfaction (β = 0.43, p < 0.000).

H4 is rejected

Non-Academic Aspects positively and less significantly affect Reputation (β = 0.15, p < 0.104).

H5 is supported

Student Satisfaction positively and significantly affects Reputation (β = 0.59, p < 0.000).

H6a is supported

Student Satisfaction mediates the relationship between Academic Aspects and Reputation (Figure 5 and Table 10).

H6b is supported

Student Satisfaction mediates the relationship between Non-Academic Aspects and Reputation (Figure 5 and Table 10).

Discussion

The purpose of this study was to analyze Taiwan’s higher education service quality in terms of student satisfaction and institutional reputation. As discussed in the literature review, all the hypotheses were developed using prior studies, none of which were conducted in Taiwan.

This study surveyed 197 international students studying in Taiwan. As mentioned in the literature review, several studies have revealed that academic aspects significantly influence reputation.^69,80 However, this study showed that different academic aspects do not significantly influence reputation. This is because of the significant and robust mediation of student satisfaction (SA). Therefore, academic aspects do not play a primary role in building a university or HEI’s reputation.

This study showed that academic aspects significantly affect student satisfaction (β = 0.40, p < 0.001). This finding confirms that students report satisfaction when universities provide suitable academic activities.^58,80,81 Students feel satisfied when universities provide high-quality academic service. Academic aspects of service quality refer to the way of teaching, and academic services refer to professors or instructors showing a positive attitude and sincere interest, behaving courteously with students, and possessing the knowledge to answer student questions. Student satisfaction results from the fulfillment of their needs in students’ minds.

This study revealed that non-academic aspects significantly affect student satisfaction (β = 0.43, p < 0.001). Alves and Raposo,⁸⁰ Hu et al.,⁸¹ Lai⁶⁰ and Subrahmanyam⁶¹ also reported the same result. The non-academic aspects of service quality play a critical role in student satisfaction. Non-academic aspects refer to office staff, administration staff, librarians, or heads of departments. International students need to adapt and adjust to a new environment and educational services. Administrative staff plays an essential role during this period of adjustment. Therefore, non-academic service quality certainly plays a role in increasing student satisfaction.

Previous studies have indicated that service quality significantly influences reputation.^65,66 However, the results of this study showed that non-academic aspects (i.e., the staff’s willingness to respond to student requests or problems, their effectiveness, their fair treatment toward students, their manner of responding toward students, and their fulfillment of student expectations) do not directly and significantly influence reputation. The insignificance of this relationship, as we will see later, is due to student satisfaction.

The literature review showed that several researchers agree that student satisfaction significantly influences reputation.^40,44 This study corroborated that student satisfaction (SS) significantly influences reputation (RP) (β = 0.59, p < 0.001). This study showed that AA significantly influences the reputation of an HEI only when mediated by SS. For international students studying in a Taiwanese HEI, SS is the most influential and essential contributor to institutional reputation. Student satisfaction significantly mediates the relationship between non-academic aspects and reputation. Therefore, student satisfaction plays an essential role as a mediator in enhancing the reputation of a university.

Conclusion and implications

This study explored the influence of both academic and non-academic aspects of service quality on international student satisfaction and institutional reputation. The results revealed that the effects of academic and non-academic aspects on reputation are insignificant. However, both academic and non-academic aspects of service quality affect student satisfaction significantly. Furthermore, the outcome revealed that student satisfaction significantly influences the reputation of HEIs. Among the three variables that affect reputation, only student satisfaction directly and significantly affects reputation. Notably, HEIs must seek to maintain or increase student satisfaction as it significantly mediates the relationship between service quality dimensions, namely, its academic and non-academic aspects, and the institution’s reputation.

Therefore, this study’s findings offer directions for HEIs in Taiwan to enhance their reputation by realizing and improving the influencing factors.

Footnotes

Appendix 1 Acknowledgments

We would like to thank Asia University for the continues support. We would like to thank Editage () for their help with formatting and English language editing of the final draft.

Declaration of conflicting interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

ORCID iD

Massoud Moslehpour

Ka Yin Chau

JingJing Zheng

References

Lapina

Roga

Müürsepp

. Quality of higher education: international students’ satisfaction and learning experience. Int J Qual Serv Sci 2016; 8: 263–278.

Chou

Ching

. Historical Overview of Education in Taiwan. In: Taiwan Education at the Crossroad. International and Development Education. New York, US: Palgrave Macmillan, 2012. DOI: 10.1057/9780230120143_3.

Statista. Taiwan: number of foreign students 2018, https://www.statista.com/statistics/1092735/taiwan-total-number-of-foreign-students (2020, accessed 24 August 2020).

Hou

Ince

Chiang

. A reassessment of Asian pacific excellence programs in higher education: the Taiwan experience. Scientometrics 2012; 92: 23–42.

Morphew

Swanson

. On the efficacy of raising your university’s rankings. In: Shin

Toutkoushian

Teichler

(eds) University rankings: theoretical basis, methodology and impacts on global higher education. Netherlands: Springer, 2011, pp.185–199.

Alvesson

Benner

. Higher Education in the Knowledge Society: Miracle or Mirage?. In: Frost

Hattke

Reihlen

(eds) Multi-Level Governance in Universities. New York, US: Springer Cham, 2016, pp.75–91. DOI: 10.1007/978-3-319-32678-8_4.

Hanaysha

Dileep

Hilman

. Service quality and satisfaction: study on international students in universities of North Malaysia. International Journal of Research in Management 2012; 3: 116–133.

Jennings

. Taiwan’s population will decline by 2021: Why that’s bad news for its tech-led economy. Forbes, https://www.forbes.com/sites/ralphjennings/2018/05/20/taiwans-population-will-decline-by-2021-why-thats-bad-news-for-its-tech-led-economy/#4c01179a5792 (accessed 17 August 2020).

Yang

. Third college closes as Taiwan’s student population shrinks by 90,000 in 10 years. Taiwan News, https://www.taiwannews.com.tw/en/news/3451996 (accessed 17 September 2019).

10.

DeAeth

. Foreign students in Taiwan 10% of total university and college students in 2018, Taiwan News. Available at: https://www.taiwannews.com.tw/en/news/3631340 (accessed 9 September 2019).

11.

Willms

Kenzie

McAllister

, et al. Effects of water quality on cattle performance. J Range Manag 2006; 55: 452–460.

12.

Churchill

Surprenant

. An investigation into the determinants of customer satisfaction. J Mark Res 1982; 19: 491–504.

13.

Parasuraman

Zeithaml

Berry

. A conceptual model of service quality and its implications for future research. J Mark 1985; 49: 41–50.

14.

Long

McMellon

. Exploring the determinants of retail service quality on the Internet. J Serv Mark 2004; 18: 78–90.

15.

Nadiri

Hussain

. Perceptions of service quality in North Cyprus hotels. Int J Contemp Hosp Manag 2005; 17: 469–480.

16.

Qin

Prybutok

Zhao

. Perceived service quality in fast-food restaurants: empirical evidence from China. Int J Qual Reliab Manag 2010; 27: 424–437.

17.

Osayawe

. Correlates of customer loyalty to their bank: a case study in Nigeria. Int J Bank Mark 2006; 24: 494–508.

18.

Paul

Mittal

Srivastav

. Impact of service quality on customer satisfaction in private and public sector banks. Int J Bank Mark 2016; 34: 606–622.

19.

Ali

Basu

Ware

. Quality measurement of Indian commercial hospitals—using a SERVQUAL framework. Benchmarking: An Int J 2018; 25: 815–837.

20.

Moslehpour

Wong

Lin

Nguyen

TLH

. Top purchase intention priorities of Vietnamese low cost carrier passengers: expectations and satisfaction. Eurasian Business Review 2018; 8(4): 371–389.

21.

Datta

Vardhan

. A SERVQUAL-based framework for assessing quality of international branch campuses in UAE. SAGE Open 2017; 7: 215824401667629.

22.

Pradela

. Quality of graduates’ preparation for labour market—A SERVQUAL analysis. Procedia Soc Behav Sci 2015; 174: 1671–1677.

23.

Ali

Raza

. Service quality perception and customer satisfaction in Islamic banks of Pakistan: the modified SERVQUAL model. Total Qual Manag Bus Excel 2015; 28: 559–577.

24.

Ahmed

Nawaz

Usman

, et al. Mediation of customer satisfaction relationship between service quality and repurchase intentions for the telecom sector in Pakistan: a case study of university students. Afr J Bus Manage 2010; 4: 34–57.

25.

Izogo

Ogba

. Service quality, customer satisfaction and loyalty in automobile repair services sector. Int J Qual Reliab Manag 2015; 32: 250–269.

26.

Brown

Mazzarol

. The importance of institutional image to student satisfaction and loyalty within higher education. High Educ 2008; 58: 81–95.

27.

Rezaei

Karami

Hajizadeh

, et al. Evaluating service quality in the higher education sector in Iran: an examination of students’ perspective. Int J Hum Rights Health 2017; 10: 146–155.

28.

Govender

Veerasamy

Noel

. The service quality experience of international students: the case of a selected higher education institution in South Africa. Mediterr J Soc Sci 2014; 5: 465–473.

29.

Cengiz

. Measuring customer satisfaction: must or not. J Nav Sci Eng 2010; 6: 76–88.

30.

Dabholkar

Shepherd

Thorpe

. A comprehensive framework for service quality: an investigation of critical conceptual and measurement issues through a longitudinal study. J Retail 2000; 76: 139–173.

31.

Hau-siu Chow

Lau

Wing-chun Lo

, et al. Service quality in restaurant operations in China: decision- and experiential-oriented perspectives. Int J Hosp Manag 2007; 26: 698–710.

32.

Oliver

. Sustainable competitive advantage: combining institutional and resource-based views. Strateg Manag J 1997; 18: 697–713.

33.

Spreng

Chiou

. A cross-cultural assessment of the satisfaction formation process. Eur J Mark 2002; 36: 829–839.

34.

Smith

Swinehart

. Integrated systems design for customer focused health care performance measurement: a strategic service unit approach. Int J Health Care Qual Assur 2001; 14: 21–29.

35.

Teas

. Expectations performance evaluations and consumers’ perceptions of quality. J Mark 1993; 57: 18–34.

36.

Hunt

. Consumer satisfaction, dissatisfaction, and complaining behavior. Journal of Social Issues 1991; 47: 107–117.

37.

Oliver

. A cognitive model of the antecedents and consequences of satisfaction decisions. J Mark Res 1980; 17: 460–469.

38.

Athanassopoulos

Gounaris

Stathakopoulos

. Behavioural responses to customer satisfaction: an empirical study. Eur J Mark 2001; 35: 687–707.

39.

Abdullah

. HEdPERF versus SERVPERF. Qual Assur Educ 2005; 13: 305–328.

40.

Ali

Zhou

Hussain

, et al.

Does higher education service quality effect student satisfaction, image and loyalty?

Qual Assur Educ 2016; 24: 70–94.

41.

Abdullah

. The development of HEdPERF: a new measuring instrument of service quality for the higher education sector. Int J Consum Stud 2006; 30: 569–581.

42.

Greimel-Fuhrmann

Geyer

. Students’ evaluation of teachers and instructional quality—analysis of relevant factors based on empirical evaluation research. Assess Eval High Educ 2003; 28: 229–238.

43.

Yeo

Thai

Roh

. An analysis of port service quality and customer satisfaction: the case of Korean container ports. Asian J Shipping Logistics 2015; 31: 437–447.

44.

Helgesen

Nesset

. What accounts for students’ loyalty? Some field study evidence. Int J Educ Manag 2007; 21: 126–143.

45.

Lahap

Ramli

Said

, et al. A study of brand image towards customer’s satisfaction in the Malaysian hotel industry. Procedia Soc Behav Sci 2016; 224: 149–157.

46.

Verčič

Žnidar

. Exploring academic reputation—Is it a multidimensional construct? Corp Commun 2016; 21: 160–176.

47.

Lievens

. Organizational image/reputation. 2nd ed. Thousand Oaks: sage, 2017.

48.

Gotsi

Wilson

. Corporate reputation: seeking a definition. Corp Commun 2001; 6: 24–30.

49.

Barney

. Firm resources and sustained competitive advantage. J Manag 1991; 17: 99–120.

50.

Rindova

Martins

. Show me the money: a multidimensional perspective on reputation as an intangible asset. Oxford Handbooks Online, 2012.

51.

Deephouse

. Media reputation as a strategic resource: an integration of mass communication and resource-based theories. J Manag 2000; 26: 1091–1112.

52.

Pfarrer

Pollock

Rindova

. A tale of two assets: the effects of firm reputation and celebrity on earnings surprises and investors’ reactions. Acad Manag J 2010; 53: 1131–1152.

53.

Rindova

Williamson

Petkova

, et al. Being good or being known: an empirical examination of the dimensions, antecedents, and consequences of organizational reputation. Acad Manag J 2005; 48: 1033–1049.

54.

Schwaiger

Raithel

. Reputation und unternehmenserfolg. Manag Rev Q 2014; 64: 225–259.

55.

Brammer

Brooks

Pavelin

. Corporate social performance and stock returns: UK evidence from disaggregate measures. Financial Manag 2006; 35: 97–116.

56.

Rhee

Haunschild

. The liability of good reputation: a study of product recalls in the U.S. automobile industry. Organ Sci 2006; 17: 101–117.

57.

Wade

Porac

Pollock

, et al. The burden of celebrity: the impact of CEO certification contests on CEO pay and performance. Acad Manag J 2006; 49: 643–660.

58.

Alves

Raposo

. Conceptual model of student satisfaction in higher education. Total Qual Manag Bus Excell 2007; 18: 571–588.

59.

Pruzan

. Corporate reputation: image and identity. Corp Rep Rev 2001; 4: 50–64.

60.

Lai

. The roles of value, satisfaction, and commitment in the effect of service quality on customer loyalty in Hong Kong–style tea restaurants. Cornell Hosp Q 2014; 56: 118–138.

61.

Subrahmanyam

. Relationship between service quality, satisfaction, motivation and loyalty. Qual Assur Educ 2017; 25: 171–188.

62.

Khoo

McGregor

. Service quality and student/customer satisfaction in the private tertiary education sector in Singapore. Int J Educ Manag 2017; 31: 430–444.

63.

Fernandes

Ross

Meraj

. Understanding student satisfaction and loyalty in the UAE HE sector. Int J Educ Manag 2013; 27: 613–630.

64.

Jung

Seock

. The impact of corporate reputation on brand attitude and purchase intention. Fashion Textiles 2016; 3: 20.

65.

Lestari

Kartini

Oesman

, et al. The impact of integrated marketing communication on corporate image and its implication to customers loyalty—case analysis at Islamic Bank in Indonesia. Eur J Bus Manag 2015; 7: 69–77.

66.

Kunanusorn

Puttawong

. The mediating effect of satisfaction on student loyalty to higher education institution. Eur Sci J 2015; 1: 449–464.

67.

Luo

Xie

Lian

. Emotional engagement and student satisfaction: a study of Chinese college students based on a nationally representative sample. Asia Pacific Educ Res 2019; 28: 283–292.

68.

Nguyen

LeBlanc

. The mediating role of corporate image on customers’ retention decisions: an investigation in financial services. Int J Bank Mark 1998; 16: 52–65.

69.

Hair

Black

Babin

, et al. Multivariate data analysis. 7th ed. New York: Pearson. 2009.

70.

Sekaran

. Research methods for business. New York: John Wiley and Sons, 2000.

71.

Jiao

. J.-P. FOX (2010) Bayesian item response modeling: theory and applications. Psychometrika 2011; 76: 360–362.

72.

Hair

Celsi

Ortinau

Bush

. Essentials of marketing research (Vol. 2). New York, NY: McGraw-Hill/Irwin, 2010.

73.

Baumgartner

Homburg

. Applications of structural equation modeling in marketing and consumer research: A review. International journal of Research in Marketing 1996; 13: 139–161.

74.

Kline

. Convergence of structural equation modeling and multilevel modeling. In: Williams

Vogt

(eds) The SAGE Handbook of Innovation in Social Research Methods. 2011, pp. 562–589.

75.

Brown

. CFA with equality constraints, multiple groups, and mean structures. In: Brown

(ed) Confirmatory factor analysis for applied research. New York, US: Guilford Publications, 2006, pp. 236–319.

76.

MacCallum

Browne

Sugawara

. Power analysis and determination of sample size for covariance structure modeling. Psychological methods 1996; 1: 130.

77.

Anderson

Gerbing

. Structural equation modeling in practice: a review and recommended two-step approach. Psychol Bull 1988; 103: 411–423.

78.

Nunnally

Bernstein

. Psychometric theory. 3rd ed. New York: McGraw-Hill, Inc., 1994.

79.

Guh

Lin

Fan

, et al. Effects of organizational justice on organizational citizenship behaviors: mediating effects of institutional trust and affective commitment. Psychol Rep 2013; 112: 818–834.

80.

Alves

Raposo

. The influence of university image on student behaviour. Int J Educ Manag 2010; 24: 73–85.

81.

Kandampully

Juwaheer

. Relationships and impacts of service quality, perceived value, customer satisfaction, and image: an empirical study. Serv Ind J 2009; 29: 111–125.